CN109863499A - Sorted out using the article that local computer tomography Distribution value is analyzed - Google Patents
Sorted out using the article that local computer tomography Distribution value is analyzed Download PDFInfo
- Publication number
- CN109863499A CN109863499A CN201680090235.1A CN201680090235A CN109863499A CN 109863499 A CN109863499 A CN 109863499A CN 201680090235 A CN201680090235 A CN 201680090235A CN 109863499 A CN109863499 A CN 109863499A
- Authority
- CN
- China
- Prior art keywords
- article
- subregion
- value
- group
- potential
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000002591 computed tomography Methods 0.000 title claims abstract description 124
- 238000009826 distribution Methods 0.000 title description 5
- 239000002360 explosive Substances 0.000 claims abstract description 37
- 238000003709 image segmentation Methods 0.000 claims abstract description 6
- 238000000034 method Methods 0.000 claims description 31
- 238000012545 processing Methods 0.000 claims description 6
- 230000007717 exclusion Effects 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 7
- 239000000463 material Substances 0.000 description 44
- 230000005855 radiation Effects 0.000 description 17
- 230000011218 segmentation Effects 0.000 description 10
- 238000007689 inspection Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 6
- 230000005611 electricity Effects 0.000 description 6
- 210000003127 knee Anatomy 0.000 description 5
- 210000004556 brain Anatomy 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000008257 shaving cream Substances 0.000 description 2
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- BUGBHKTXTAQXES-UHFFFAOYSA-N Selenium Chemical compound [Se] BUGBHKTXTAQXES-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000003796 beauty Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 239000011469 building brick Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000005059 dormancy Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 229910052711 selenium Inorganic materials 0.000 description 1
- 239000011669 selenium Substances 0.000 description 1
- -1 shirt Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
- G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
- G01N23/046—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material using tomography, e.g. computed tomography [CT]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V5/00—Prospecting or detecting by the use of ionising radiation, e.g. of natural or induced radioactivity
- G01V5/20—Detecting prohibited goods, e.g. weapons, explosives, hazardous substances, contraband or smuggled objects
- G01V5/22—Active interrogation, i.e. by irradiating objects or goods using external radiation sources, e.g. using gamma rays or cosmic rays
- G01V5/226—Active interrogation, i.e. by irradiating objects or goods using external radiation sources, e.g. using gamma rays or cosmic rays using tomography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/162—Segmentation; Edge detection involving graph-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2223/00—Investigating materials by wave or particle radiation
- G01N2223/40—Imaging
- G01N2223/419—Imaging computed tomograph
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/05—Recognition of patterns representing particular kinds of hidden objects, e.g. weapons, explosives, drugs
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Pulmonology (AREA)
- Immunology (AREA)
- Pathology (AREA)
- High Energy & Nuclear Physics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
In addition, the present invention provides one or more systems and/or technology for being sorted out to the article being placed in object.The three-dimensional image segmentation of object (for example, bag) is indicated into (for example, laptop computer, thermos bottle etc.) at one group of article.Article characteristics based on article, such as can be used for hiding the laptop computer of the item of interest of such as explosive, article is identified from this group of article expression.The region of 3-D image including the article is divided into one group of subregion (for example, the first subregion comprising screen, second subregion comprising motherboard etc.).When the number of voxel in the range of known CT value of computer tomography (CT) value in first kind article in any subregion is more than threshold value, the article is classified as potential first kind article (for example, explosive laptop computer).
Description
Background technique
This application involves the fields computer tomography (CT).Present invention discover that using the specific application of security system, it should
Using being used to that object be imaged and identify the item of interest in the object, such as potential threat article.This hair
It is bright to further relate to medicine, industry and/or other application, in such applications, identify that the article in object to be checked will be useful.
Safety is major issue at airport and other travelling relevant ranges.It is a kind of for promoting the skill of touring safety
Art is baggage check.In general, being convenient for baggage using radiant image mode.For example, CT system can be used for Security Officer
The two dimension and/or 3-D view of object are provided.After checking the image provided by imaging device, Security Officer can determine to go
Lee whether safety security checkpoints or whether it is necessary to carry out further it is (manual) inspection.
For the associated mistake of the potential threat article for reducing with identifying in the luggage, can use automatic
Change object recognition system.This system can extract the shaving cream container in article, such as bag, and base from the image of object
Goods attribute is usually calculated in attribute/member of image.Then, by by goods attribute (for example, density, effective atomic number, shape
Shape etc.) it is compared with the known attribute of menace article, non-threatening items and/or the two classifications etc., institute can be used
Calculated goods attribute distinguishes the article.
Regrettably, the artifact in images of items may be decreased described in the correct identification of the Automation object identifying system
The article marking is threat or non-threat ability by article.These artifacts are usually generated by high density assembly, such as electricity
Component (for example, CD-ROM drive, Notebook Battery or other high atomic numbers Z component) inside sub- equipment, and described image can be made
Characteristic (for example, density value, z- virtual value etc.) distortion.Therefore, it is identified and is hidden in these equipment using professional technique
Menace article (for example, explosive).
Summary of the invention
The various aspects of the application solve the above problem and other problems.According on one side, provide it is a kind of for pair
The method that the article being placed in object is sorted out.The method includes receiving the 3-D image of the article and by the object
The 3-D image of product is divided into one group of subregion.The method also includes: for the first subregion in this group of subregion,
Computer tomography (CT) value based on each voxel being arranged in the first subregion, will be arranged in the first subregion
Voxel binning is to first group of histogram every in (bin).The method also includes in first subregion and CT value exists
When the number of voxel in the range of the known CT value of first kind article is more than specified threshold, the article is classified as potential
First kind article.
A kind of calculating equipment is provided according to another aspect, which includes processing unit and memory.It is described to deposit
Reservoir includes processor-executable instruction.The processor-executable instruction executes operation when being executed by the processing unit.It should
Operation includes indicating the three-dimensional image segmentation of object at one group of article.The operation further includes the object for calculating this group of article and indicating
Product feature.Operation further includes the article indicated based on the potential hiding article for potential hiding article in this group of article expression
Feature identifies potential hiding article from this group of article expression.Operation further includes limiting to pair comprising the potential hiding article
Region in the 3-D image of elephant.It is one group of subregion that operation, which further includes by the region division,.Operation further includes when first
In subregion and when voxel number of CT value in the range of the known CT value of first kind article is more than specified threshold, by this
Potential hiding article is classified as including potential first kind article.
A kind of non-transitory computer-readable medium is provided according to another aspect, the non-transitory computer-readable medium packet
Include computer executable instructions.The computer executable instructions execute operation when being executed by processing unit.The operation includes will
The three-dimensional image segmentation of object is indicated at one group of article.Operation further includes the article characteristics for calculating this group of article and indicating.Operation is also
Including the article characteristics indicated based on the potential hiding article for potential hiding article in this group of article expression, from this group of object
Potential hiding article is identified in product expression.Operation further includes the area in the 3-D image for limit the object comprising potential hiding article
Domain.It is the first subregion, the second subregion and third subregion that operation, which further includes by the region division,.Operation further includes assessment
The CT value of first subregion, the second subregion and the voxel in third subregion exists with the example for determining first kind object
In in any one of the first subregion, the second subregion or third subregion subregion.Operation further include when determine this
When one type object is present in any one of the first subregion, the second subregion or third subregion subregion, by this
Potential hiding article is classified as including potential first kind article.
After specification appended by reading and understanding, those of ordinary skill in the art will appreciate that its other party of the application
Face.
Detailed description of the invention
The application is shown in the accompanying drawings by way of example, and not limitation.In the accompanying drawings, identical appended drawing reference is usual
Indicate similar element, and wherein:
Fig. 1 shows the example context of imaging pattern.
Fig. 2 shows the example contexts for threatening determiner.
Fig. 3 shows the example of segmentation 3-D image.
Fig. 4 shows the example for identifying potential hiding article.
Fig. 5 shows the example for limiting the region of 3-D image of the object comprising potential hiding article.
It is showing for one group of subregion that Fig. 6, which is shown the region division of the 3-D image of the object comprising potential hiding article,
Example.
Fig. 7 shows the example of computing computer tomography (CT) value.
Fig. 8 shows the example sorted out to potential hiding article.
Fig. 9 shows the flow chart of the illustrative methods for being sorted out to the article being placed in object.
Figure 10 shows the computer readable media including processor-executable instruction, and the processor is executable
Instruction is configured as implementing one or more configurations set forth herein.
Specific embodiment
Theme claimed is described referring now to the drawings, wherein identical appended drawing reference is commonly used in always showing phase
Same element.In the following description, for illustrative purposes, numerous specific details are set forth so as to theme claimed
It is understood thoroughly.It will be apparent, however, that can practice without these specific details claimed
Theme.In other cases, it is shown in block diagram form structure and equipment, in order to describe theme claimed.
In addition, there is provided herein for being sorted out to the article being placed in object one or more systems and/or
Technology.Object (such as bag) may include various articles, such as shoes, laptop computer, books etc..One of article can
With the interested article for hiding such as menace article (for example, explosive).For example, the menace article can be hidden
Be hidden in laptop computer (for example, hard disk drive can be removed from the hard drive bay in laptop computer, and
Explosive can be stored in the hard drive bay).In order to assess bag to identify item of interest, can scan
The bag is to generate the 3-D image of bag.One group of article to identify the article in bag can be split to 3-D image
Indicate (for example, wherein each article indicates it is the three-dimensional figure for describing the single item of such as laptop computer in the bag
Picture).The attribute (for example, density, effective atomic number, shape etc.) of each article can be extracted from article expression, and
And can be compared the attribute of article with the known attribute of menace article, to determine whether the article is potentially to threaten
Property article.Regrettably, when menace material is hidden in certain form of article such as electronic device or other are flat highly dense
When spending in object, it is possible that failing to report, this is because indicating that the adjoining of the voxel of menace material may be by image artifacts
It is broken to the part of multiple disconnectings.In addition, if the menace material is placed on the hiding article of such as laptop computer
In interior multiple adjacent but different region, then it will appear and fail to report.
Therefore, as herein provided, hiding for such as electronic product can be hidden in identify with the accuracy of raising
The item of interest of such as menace article in article.Specifically, identification includes the three-dimensional of article (for example, laptop computer)
Then the region division is one group of subregion by the region of image.E.g., including the region quilt of the 3-D image of laptop computer
It is divided into the first subregion of the screen comprising laptop computer, the second subregion of motherboard comprising laptop computer, includes
The third subregion of hard drive bay and laptop computer other assemblies etc..Subregion can be any shape, size or match
It sets.Subregion may include entire component (for example, battery), a part of component or its multiple component or part.In addition, sub-district
Domain can be (for example, not the sharing voxel between subregion) of mutual exclusion or subregion and can be overlapped (for example, some subregions can
To share at least some voxels).
Subregion is assessed based on single subregion to determine whether there is certain types of article in subregion, such as threatened
Property material.For example, can the CT value (for example, density value, z virtual value etc.) based on each voxel being arranged in subregion will be sub
Voxel binning in region is to one group of histogram in.When in subregion and CT value a kind of types of articles known CT value
In the range of the number of voxel exceed a prescribed threshold value (for example, independently of the CT of the voxel in other subregions of the article
Value) when, article can be classified as to potential the type article (for example, menace article).In this way, when electricity on knee
When one or more subregions of brain or other articles may include explosive, laptop computer can be classified as to explosive object
Product.
Fig. 1 shows the exemplary environments 100 including computer tomography (CT) system, and the CT system can be matched
It is set to the image for generating the one or more aspects for indicating object 102 (for example, luggage, patient etc.) or object in inspection,
And the certain form of article in object 102 is hidden in using one or more image detections generated.It is such exemplary
System can be used in the luggage that example is imaged as has already been identifying from a kind of article potentially containing threaten article (for example,
The menace substance being hidden in the potential hiding article of such as electronic equipment or thin Dense objects).
It is understood that although exemplary environments 100, which describe, is configurable to generate in inspection (or checking before)
The two dimension of object 102 and/or the CT system of 3-D image, but it is also contemplated that other radiophotography modes are for generating figure
Picture can detect and sort out to be used for come the article in cutting object 102 according to the image.In addition, in the exemplary environments 100
Including the arrangement of component and/or the type of component only exemplary arrangement is used as to provide.For example, in some embodiments, number
It is included in detector array 106 according to acquisition component 122.
In the exemplary environments 100, the inspection equipment 108 of CT system is configured to check for the one of such as object 102
A or multiple objects.Check that equipment 108 may include rotary frame 104 and (fixation) support construction 110 (for example, it can wrap
Firmly and/or around the rotary frame 104 at least part (for example, as the outer peripheral outside around internal rotating ring is solid
Determine shown in ring)).During check object 102, object 102 can be placed on supporting element 112, such as selectively locate
The bed or conveyer belt of (for example, hollow hole in rotary frame 104) in inspection area 114, and rotary frame 104 can be with
It rotates and/or supports around object 102 by rotator 116 (such as motor, drive shaft, chain, roller steering frame etc.).
Rotary frame 104 can surround a part of inspection area 114, and may include one or more radiation sources
118 (for example, ionization x-ray source, gamma emitters etc.) and detector array 106.The detector array is mounted on rotary frame
104 relative on the substantially radial opposite side of radiation source 118.
During check object 102, radiation source 118 is from the focus of radiation source 118 (for example, in the radiation for issuing radiation 120
Region in source 118) radiation 120 of transmitting sector, taper, wedge shape and/or other shapes configures to the inspection area 114.
It is appreciated that can essentially continuously emit and/or can intermittently emit (for example, stopping what radiation source 118 was not activated
Emit of short duration impulse radiation after the dormancy phase) this radiation 120.
As the radiation 120 emitted passes through object 102, radiation 120 can be different due to the different aspect of object 102
Ground decaying.The radiation 120 of different weight percentage because different aspect can decay, it is possible to based on decaying or by detector array
The variations of 106 photon numbers detected generates image.For example, less intensive compared to such as skin or clothes of object
Aspect, such as bone of object 102, metal plate, electronic component more dense aspect can decay more radiation 120
(for example, leading to less photon strikes detector array 106).
Detector array 106 be configured as the radiation that will test directly convert (for example, using amorphous selenium and/or other
Direct transition material) and/or indirect conversion (for example, using photodetector and/or other indirect conversion materials) be signal, detection
Device array 106 can send the signal to data acquisition components 122.Data acquisition components 122 are configured with various skills
The signal that art (for example, integral, photon counting etc.) compiling is sent at preset time intervals or in measurement interval.It is understood that
It is that such measurement interval can be referred to as " view " and usually react to be in from radiation source 118 relative to object 102
The signal that the radiation 120 emitted when special angle range generates.Based on the compiling signal, data acquisition components 122 be can be generated
Such as the data for projection of instruction compiling signal.
The exemplary environments 100 further include image reconstructor 124, are configured as receiving by data acquisition components 122
The data for projection of output.Image reconstructor 124 is configured with suitable analysis, iteration and/or other reconstruction technique (examples
Such as, backprojection reconstruction, chromatographical X-ray synthesis reconstruction, iterative approximation etc.) it is generated according to data for projection to the three-dimensional figure as 102
As data (also referred to as 3-D image).In this way, data are transformed into image space from projector space, for example, can make
The domain that the user 134 of observation image is more readily understood.
In exemplary environments 100, Object Segmentation component 126, which can be configured as, to be received image data and divides the figure
As the article indicated in data.For example, the image of suitcase can describe clothes, hair dryer, beauty production in suitcase
Product, laptop computer and/or other articles, and the voxel of image data can be divided into its group by Object Segmentation component 126
At part (for example, by the voxel for the image data for representing laptop computer and the voxel for the image data for representing hair dryer point
It opens).In this way it is possible to be individually separated and analyze the expression of the article in suitcase.
In exemplary environments 100, threaten determiner 128 can from image reconstructor 124 receive image data and/or from
Object Segmentation component 126 receives image data.As will be described in more detail, threaten determiner 128 can be configured as by
Indicate that the image data (for example, dividing via Object Segmentation component 126 according to the image data of object) of article is divided into
One group of subregion (for example, being one group of subregion by the region division of the 3-D image including the laptop computer).It threatens true
Determining device 128 can be additionally configured to individually assess each subregion to determine whether subregion includes menace material.For example, prestige
Side of body determiner can assess whether have the CT value of enough voxels in the certain kinds for being directed to such as menace article in subregion
In the range of the known CT value of type article, and correspondingly laptop computer is sorted out.
The exemplary environments 100 further include terminal 130 or work station (for example, computer), can be configured as and connect
Receipts image data (for example, exported by Object Segmentation component 126 and/or image reconstructor 124) and/or by threat determiner
The information (for example, alarm that laptop computer may include menace material) of 128 outputs.Terminal 130 can be additionally configured to
To (for example, the Security Officer, medical worker etc.) display image data of user 134 and/or information on monitor 132.With this
Mode, user 134 can be with check images to identify that area-of-interest in object 102 and/or reception detect menace article
Alarm.Terminal 130 can be additionally configured to receive user's input, and user's input can for example indicate to check the behaviour of equipment 108
Make (for example, speed of conveyer belt).
In the exemplary environments 100, controller 136 is operatively coupled to terminal 130.In one example, it controls
Device 136 processed is configured as example receiving the input of such as user input of self terminal 130, and generates for checking equipment
The instruction of 108 instruction on-unit.For example, user 134 may want to reexamine object 102, and controller 136
Order instruction supporting element 112 can be issued and turn direction (for example, object 102 to be brought back to the inspection area for checking equipment 108
114)。
It is appreciated that example components figure is merely intended to show one embodiment of a type of imaging pattern, and not
It is explained in a manner of limitation.For example, the function of one or more components described herein can be assigned to multiple components, and/or
The function of two or more components described herein can be incorporated into only single component.In addition, the imaging pattern can wrap
Including the add-on assemble for being configured as executing supplementary features, function etc. and/or some components described herein can be optionally.
Fig. 2 is showing for the exemplary threat determiner 128 sorted out for using Local C T Distribution value to analyze to article
Meaning property block diagram 200.Object Segmentation component 126 can receive the 3-D image 201 of the object of such as bag, such as the example 300 of Fig. 3
Shown in.The 3-D image 201 can be divided into one group of article of the article in bag to indicate 202 by Object Segmentation component 126
(for example, 3-D image of each article in bag), such as laptop computer indicate that 206, books indicate that 306, shoes indicate
308 etc..In the example of segmentation 3-D image 201, density feature, effective atomic number, size characteristic, shape feature etc.
Article characteristics can be used to identify edge between article and/or isolation and indicate the voxel of different articles, it is described above-knee to create
Type computer indicates that the 206, books indicate 306 and shoes expression 308 etc..
The entry control assembly 204 of determiner 128 is threatened to can be configured as article of special interest in identification object.
For example, entry control assembly 204 can be configured as the certain types of article (for example, electronic product) of identification, it is specific according to this
The article of type, the specific application of technique described herein discovery.As an example, with reference to the example 400 in Fig. 4, entry control
Component 204 can be configured as the image data of the one or more articles of analysis (for example, what segmentation obtained), based on basis point
The article in the object of shape, density or z effective information that image obtains to identify particular category is cut (for example, electronic product, thin
High density article etc.).Using the analysis, entry control assembly 204 can indicate that 206 article is special based on the laptop computer
Sign is identified as laptop computer expression 206 to describe potential hiding article (for example, article of special interest).With this
Laptop computer expression 206 is identified as the 3-D image of potential hiding article, is used for into one by mode, entry control assembly 204
Step assessment article sorts out and/or threat detection.
Although it is appreciated that it is referenced be the subset of all items in object 102 is executed threat detection (for example,
Only those of potential hiding article article execution threat detection is identified as to by entry control assembly 204), this description and unawareness
Threat detection is not executed to other articles in object 102 in implying.On the contrary, the technology except scope of the present application can be used
To identify and/or sort out those articles.In yet another embodiment, each article in object 102 can be executed and is retouched here
The technology stated, therefore the entry control assembly 204 can be not provided with.
Referring back to Fig. 2, such as described laptop computer can be indicated to the image of 206 article of special interest
(for example, image of potential hiding article) is sent to region limitation unit 208.Region limitation unit 208 is configured as passing through packet
Voxel adjacent with the article in 3-D image is included to extend the 3-D image of the article (for example, creating around the article slow
Area is rushed, to ensure that entire article is included in the 3-D image of the article).
In some embodiments, the coordinate system from restriction relative to the article carrys out starting region limitation unit 208.The coordinate
System can be independently of to image orientation or dependent on to image orientation.As the example independently of the coordinate system to image orientation, region
Limitation unit 208 can determine the European vector limited by inspection equipment 108 (for example, the Z axis of article and supporting element 112 are mobile
The direction of article is corresponding).As the example dependent on the coordinate system to image orientation, region limitation unit 208 can be determined and is based on
The feature vector that the size of the article defines.As an example, region limitation unit 208 has fixed with reference to the example 500 in Fig. 5
The feature vector of justice.First eigenvector 502 (for example, X-axis) is defined along the long axis of laptop computer.Accordingly, second feature
Vector 504 (for example, Y-axis) and third feature vector 506 (for example, Z axis) are defined as perpendicular to first eigenvector 502.
Using the coordinate system, region limitation unit 208 limits the region of the 3-D image 201 of the object comprising the article
210 and the voxel adjacent with the voxel for representing the article.For example, region limitation unit 208 can be set to there are two tools
The minimum threshold of distance of voxel.Therefore, region limitation unit 208 can form the region around laptop computer expression 206
210, the region include indicate laptop computer voxel, close to those indicate laptop computers voxels voxel, Yi Jiyu
Indicate that the voxel of laptop computer is at most spaced apart those of voxel voxel.Therefore, the region 210 may include indicating
The voxel of laptop computer, and further include around the buffer area of laptop computer, which includes not representing electricity on knee
The voxel of brain (represents air, clothes or any other article in the near space close to laptop computer for example, being changed to
Voxel).
Region limitation unit 208, which can be configured as, in yet other embodiments, is defined with preboarding for region 210
Shape.The size of the predetermined shape is the function of minimum threshold of distance and coordinate system.For example, region limitation unit 208 can be with
It is configured as limiting the box regions 210 for surrounding the laptop computer, and at least ensures that and represent the laptop computer
Voxel is surrounded by the buffer area of at least two voxels.Using the standard, region limitation unit 208 can be created around electricity on knee
Brain indicates 206 box regions 210, and the size of the box regions is the function of aforesaid standards.It is understood that due to the mark
Quasi- and article shape is spaced apart the voxel of three or more voxels with those of article voxel is indicated in some regions
It may include in box regions 210.
In other embodiments, the region is defined as only including indicating those of article voxel and indicating article
Voxel preset distance in those of voxel.Therefore, the shape in region can substantially match with the shape of article.
Referring back to Fig. 2, threaten the subregion limitation unit 212 of determiner 128 that can will be limited by region limitation unit 208
Fixed region 210 is divided into one group of subregion.In some embodiments, the seat limited by region limitation unit 208 can be used
How region 210 is divided into one group of subregion to limit by mark system.
As an example, the region 210 can be divided by subregion limitation unit 212 with reference to the example 600 in Fig. 6
Multiple subregions 214, wherein each subregion 602,604 and 606 is indicated along coordinate vector (for example, second feature vector
504) layer (such as piece) in the region 210 stacked.Therefore, each subregion 602,604 and 606 includes indicating laptop computer
A part of voxel and the voxel including the buffer area adjacent with the voxel of a part of laptop computer is indicated in this layer.
For example, the region 210 comprising laptop computer can be divided into the first subregion 602 comprising lap top screens, packet
Second subregion 604 of keyboard and hard drive bay containing laptop computer is (for example, wherein stored menace material
608) and the third subregion 606 comprising laptop computer mainboard material.
It is appreciated that any amount of subregion can be limited, and subregion can have any shape, size or match
It sets.In addition, the quantity of subregion and/or their shape, size or configuration can be based on the classes of the potential hiding article of analysis
Type.For example, when the potential hiding article of analysis is camera selected subregion shape and/or size can with when potential
The shape and/or size of selected subregion are different when hiding article is laptop computer, this is because difference in shape
And/or about the position for the explosive materials that can be hidden in article and the priori knowledge of degree.
In addition, the subregion can be (e.g., including the voxel of non-overlap) of mutual exclusion or can partly be overlapped
(for example, at least some voxels can be shared between the subregion).In addition, subregion can have identical or different thickness
Degree, this can type and/or potential hiding article based on the potential hiding article in analysis component attribute (for example, main
The shape and size of plate).Further, subregion can be defined as the entire component comprising article (for example, entire above-knee
Type computer screen), a part (for example, a part of lap top screens) of component or multiple components be (for example, electricity on knee
A part of brain screen and a part of keyboard).
Referring back to Fig. 2, threatens the CT value of determiner 128 to determine that component 216 can be configured as and determine in this group of sub-district
The CT value of each subregion in domain 214, or can be configured as the distribution for determining these CT values.As it is used herein,
CT value can be calculated density value and/or z virtual value in the 3-D image of 124 reconstructed object of image reconstructor.Citing
For, each voxel in subregion can with voxel-based CT value by a histogram of the binning to one group of histogram in every
In, wherein each histogram is every the one or more CT values that can correspond to different range.
With reference to Fig. 7, the example 700 that description CT value determines the operation of component 216 is shown.The CT value determines 216 quilt of component
It is configured to create one group of histogram 218, each histogram 712,716 and 720 in this group of histogram 218 for region 210
Indicate a sub-regions 602,604 and 606.For example, CT value determines that component 216 can be for the first subregion 602 creation first
Histogram 712 creates the second histogram 716 for the second subregion 604, and creates third histogram for third subregion 606
Figure 72 0.The X-axis 704 of each histogram 712,716 and 720 can indicate CT value, and each histogram 712,716 and 720
Y-axis 702 can indicate the number of voxel.Therefore, histogram 712,716 and 720 can show for each subregion and have together
The number of the voxel of one CT value.
Although in other embodiments, histogram can be item it is appreciated that exemplary histograms depict line chart
Shape figure, wherein each bar shaped indicates the range (for example, each bar shaped represents more than one CT value) of CT value.In addition, histogram every
Size (for example, the number for the CT value that bar shaped indicates) can be different on the histogram.For example, explosive may fall in it is specified
Within the scope of CT.Indicate some histograms of the CT value far from specified CT range every can be bigger (for example, each histogram is every can be with
Indicate the CT value of 100 unit ranges), and indicate histogram close to the CT value of specified CT range every can be smaller (for example, every
A histogram is every the CT value that can represent 30 unit ranges), for example to better discriminate between explosive and its CT value near-by explosion
The benign article of the CT value of object.
Referring back to Fig. 2, threatens the classification component 220 of determiner 128 can be configured as and receive this group of histogram 218 simultaneously
The each histogram for assessing this group of histogram 218 will be analyzing article and be classified as potential first kind article (example
Such as, menace article) or Second Type article (for example, non-threatening items or benign article).For example, classification component 220 can
With assess the histogram that is obtained from it every or histogram, indicated and the CT value pair of explosive with being determined for any subregion
Whether the histogram for the CT value answered is more than specified threshold every the number of interior voxel.If in one or more subregions,
The number of voxel in the range of given CT value is more than the number of specified threshold or voxel in the range of given CT value
Specified threshold is had exceeded relative to expected number, then article can be classified as to menace article.
For example, with reference to the example 800 in Fig. 8, the reception of classification component 220 is determined by the CT value in example 700
One group of histogram 218 that component 216 generates, and the histogram of each subregion is individually assessed, with determine whether should be by knee
Laptop is classified as potential first kind article, such as the menace article comprising menace material.When in subregion
And body of the CT value in the range (for example, for range of the known CT value of explosive) of the known CT value of first kind article
When the number of element has been more than specified threshold, laptop computer can be classified as to potential first kind article (for example, potential
Explosive articles).
For example, the classification component 220 can by for the first subregion 602 the first histogram 712 in CT value 710 with
It is compared for the expection CT value 706 in the histogram 708 of the expection CT value for the laptop computer for not having menace material.
Because the expection CT value 706 that the CT value 710 in the first histogram 712 deviates in histogram 708 is not above specified threshold,
It can determine that the first subregion 602 includes the desired material of the laptop computer without menace material.For example, the first histogram
The expection CT value peak value 709 that the peak value 713 of CT value in 712 deviates in the histogram 708 of expected CT value is not above threshold quantity.
Accordingly, it can be determined that the first subregion 602 includes the desired material of the laptop computer without menace material.
Classification component 220 can by for the second subregion 604 the second histogram 716 in CT value 714 be directed to do not have
There is the expection CT value 706 in the histogram 708 of the expection CT value of the laptop computer of menace material to be compared.Because second
The threshold quantity of CT value 714 in histogram 716 deviates from the expection CT value 706 in histogram 708, so can determine the second sub-district
Domain 604 includes the material of the first kind, such as explosive materials or other menace materials.Although for example, the second histogram 716
The expection CT value peak value 709 that the peak value 715 of interior CT value deviates in the histogram 708 of expected CT value is not above threshold value, still
Second peak value 717 of the CT value in the second histogram 716 deviates the expection CT value peak value 709 in the histogram 708 of expected CT value
It has been more than threshold quantity.Hence, it can be determined that the second subregion 604 includes being not intended to the material being included in laptop computer.
In addition, second peak value 717 may be fallen within the scope of the expection CT of menace material.It is therefore found that second subregion
604 may include menace material.In this way, whether other subregions regardless of laptop computer have indicated menace material, it will
The laptop computer is classified as potential threat article (for example, potential first kind article) 222.
In some embodiments, once identifying the type (for example, menace material) of instruction item of interest such as
When the exception of the second peak value 717, program can stop (for example, the CT value determines that component 216 can stop assessing subregion simultaneously
And it is necessary to further (for example, manually) check for notice terminal 130).In other embodiments, though a sub-regions
It is identified as to can continue to analyze other comprising interested material (for example, explosive materials), classification component 220
Subregion.For example, laptop computer is returned even if classification component 220 is determining due to the exception in the second subregion 604
Class is potential threat article, and classification component 220 can also will be directed to the CT value in the third histogram 720 of third subregion 606
718 carry out with the expection CT value 706 in the histogram 708 for the expection CT value of the laptop computer without menace material
Compare.Because the CT value 718 in third histogram 720 deviates the expection CT value 706 in histogram 708 without departing from specified threshold
Value, it is possible to determine that third subregion 606 includes the desired material of the laptop computer without menace material.For example,
Expection CT value peak value 709 in the histogram 708 that the peak value 719 of the CT value in third histogram 720 deviates expected CT value does not have
Have more than threshold quantity.Therefore, third subregion 606 can be determined as including the pre- of the laptop computer without menace material
Phase material.
Because can be sorted out based on single subregion to laptop computer, it is possible to regardless of menace material is
It is no visually discontinuous or whether physically discontinuous and identify interested material (for example, menace material 608),
This is visually discontinuously attributed to the artifact for visually splitting the 3-D image 201 of single menace article, and does not connect physically
The continuous menace material that is attributed to is physically divided into different lesser threat articles, these lesser menace articles are usual
Not comprising the enough menace materials or size for being identified as threatening.
Fig. 9 shows the illustrative methods 900 for being sorted out to the article being placed in object.More specifically, institute
It states illustrative methods 900 and describes the technology of potential first kind article (for example, explosive) for identification, which can
Can it is due to the artifact in the 3-D image of article and visually discontinuous (for example, explosive can be single physical article,
But the artifact of three-dimensional article may cause the explosive and look like multiple lesser explosives, the lesser explosive
To not have and be individually classified as the enough menace materials or size of menace object) or the article due to being stored in
It is physically discontinuous (for example, explosive, which can be physically separated into, is stored in camera in the individual region of article
The lesser explosive of interior different location).
The illustrative methods 900 start from 902, and at 904 by the three-dimensional image segmentation of object at one group of article table
Show.For example, 3-D image may include the CT image for describing bag object.The bag object may include one group of article, such as
Shaving cream, shirt, tablet computer and/or other articles.At 906, the article characteristics of this group of article expression can be calculated.Article
Feature may include density feature, shape feature, dimensional characteristics, atom value tag of this group of article etc..For example, table can be based on
The shape feature and density feature of the bright thin close article including electronic building brick identifies tablet computer.
At 908, potential hiding article can be identified based on article characteristics.For example, the electricity of such as tablet computer etc
Sub- product may be potential hiding article, this is because due to electronic product is flat intensive object and is difficult in electronics
The interested article of such as menace material (for example, explosive) is detected in product, thus electronic product can be used for it is hidden
Hide interested article.In this way it is possible to which there is the article characteristics (example for indicating electronic equipment based on potential hiding article
Such as, density, atomic number, shape, size etc.) potential hiding article indicate (for example, segmented image of potential hiding article)
To identify the potential hiding article of such as tablet computer.
At 910, the region in the 3-D image of object comprising the potential hiding article can be limited to.The region can be with
Voxel (for example, voxel that potential hiding article indicates) comprising indicating potential hiding article.It the region can also be including not table
Show the voxel (for example, in 3-D image of object, the adjacent voxels adjacent with potential hiding article) of potential hiding article
Buffer area.For example, buffer area may include save the tablet computer tablet computer set some voxels and in bag with
The voxel of the adjacent shirt of tablet computer.The buffer area can be defined as including any number for not indicating the potential hiding article
The voxel of amount, shapes or configure.The region may include any shape, size or configuration.
It can be one group of subregion by the region division at 912.This group of subregion may include potential hiding article
Layer (piece) (for example, each layer may include some voxels for indicating potential hiding article and some voxels in buffer area),
These layers obtain for example along the dimension of 3-D image along the feature vector determined for potential hiding article.These subregions can
With identical or different thickness.In addition, these subregions can be mutual exclusion (for example, between two sub-regions not altogether
Enjoy voxel) or overlapping (for example, some voxels can be shared between two sub-regions).
For first subregion of this group of subregion, each voxel being arranged in first subregion can handle
CT value, to sort out to potential hiding article.The CT value can correspond to density information, z effective information or any other
Data.In this example, can based on CT value by voxel binning at one group of histogram in.The histogram is special every that can indicate to have
Determine the number of the voxel of the CT value of range.In another example, which can be used to generate the histogram (example of CT value
Such as, local density's distribution of subregion).The histogram can draw the number of the voxel with specific CT value.
At 918, when in the first subregion and CT value first kind article known CT value (for example, explosive
CT value) in the range of voxel number be more than specified threshold when, potential hiding article can be classified as including potential
First kind article (for example, explosive tablet computer).In this example, if range with the known CT value of first kind article
Corresponding histogram is more than threshold value every the number of interior voxel, then can determine that the first subregion includes first kind article
Material, such as explosive materials.In this way it is possible to which the potential hiding article of such as tablet computer is determined as explosive object
Product.
It is appreciated that each subregion in this group of subregion can individually be assessed to return to potential hiding article
Class.In this example, it can determine that first subregion and second subregion do not include in first kind article
The voxel of number of thresholds within the scope of known CT value.However, it is possible to determine that third subregion includes in first kind article
The voxel of number of thresholds within the scope of known CT value.In this way, even if determining that the first subregion and the second subregion do not wrap
The material of first kind article is included, tablet computer can also be classified as to the potential first kind article, such as explode mild-natured
Plate computer.
In another example, potential hiding article can be classified as first kind article, although the first kind object
Product be visually it is discontinuous (for example, the artifact of 3-D image may cause single continuous explosive appear as compared with
Small explosive, the lesser explosive do not include the enough explosive materials or size for being classified as explosive)
It or is physically discontinuous (for example, explosive may physically be divided into lesser single explosive and be stored in this
The different location of tablet computer).For example, it may be determined that the first subregion and third subregion (for example, discontinuous subregion) packet
Include the voxel of the number of thresholds within the scope of the known CT value of first kind article.However, it is possible to determine the second subregion (example
Such as, adjacent with the first subregion and third subregion) it does not include the threshold within the scope of the known CT value of first kind article
It is worth the voxel of quantity.In this way, it even if the first subregion and third subregion are not adjacent subregions, can also will put down
Plate computer is classified as the potential first kind article, such as explosive tablet computer.
The illustrative methods 900 terminate at 920.
Another embodiment is related to a kind of computer-readable medium including processor-executable instruction, and the processor can be held
Row instruction is configured as realizing one or more technologies presented herein.Show in Figure 10 to design in such ways
Computer readable media.Implementation method 1000 includes having the calculating of the mechanized data 1004 of coding on it
Machine readable medium 1002 (for example, disc of CD-R, DVD-R or hard disk).The mechanized data 1004 includes at one group again
Device executable instruction 1006 is managed, is configured as being operated according to one or more principles set forth herein.In this way at one
Embodiment 1000 in, processor-executable instruction 1006 can be configured as execution method 1008, such as the exemplary side of Fig. 9
At least some of method 900.In embodiment as another, processor-executable instruction 1006 can be configured as realization
System, such as at least some of the exemplary environments 100 and/or at least some of the system 200 of Fig. 2 of Fig. 1.This field is general
Logical technical staff can design many such computer-readable mediums, be configured as according to given herein one or more
Technology is operated.
It is appreciated that " example " and/or " exemplary " is used herein to mean that as example, example or explanation.Herein
In be described as any aspect, the design of " example " and/or " exemplary " etc. be not necessarily to be construed as than other aspect, design etc. more
Favorably.On the contrary, the use of these terms is intended to that concept is presented in specific ways.As used in this specification, term "or"
It is intended to indicate that the "or" of inclusive rather than exclusive "or".That is, it is unless otherwise indicated or clear from the context,
Otherwise " X uses A or B " is intended to indicate that any natural inclusive arranges.That is, using A, X to make using B or X in X
With in any example of both A and B, all meet " X uses A or B ".In addition, being preced with used in the application and appended claims
Word " one " and "one" can usually be construed to indicate " one or more ", be clearly directed toward unless otherwise indicated or from context
Singular.In addition, at least one of A and B etc. usually indicates A or B or both A and B.
Although the present invention, those skilled in the art have shown and described relative to one or more implementations
Equivalent change and modification will be expected based on the reading and understanding to the specification and drawings.The present invention includes all such modifications
And variation, and only limited by scope of the appended claims.Particularly, about by said modules (for example, element, resource etc.)
The various functions of executing, unless otherwise stated, the term for describing these components, which is intended to correspond to, executes the component
Specified function any component (for example, functionally equivalent), even if the component is not equal in structure of the invention
The disclosed structure of function is executed in example shown here implementation.Similarly, the sequence of shown movement is simultaneously
It is not intended to be restrictive, so that the different sequences including the same action in different (for example, quantity) movement are intended to fall within
In the scope of the present invention.In addition, although the present invention may be disclosed only about the only one implementation in several implementations
Special characteristic, but such feature can be expectation and other advantageous realization sides for any given or specific application
One or more other features combinations of formula.In addition, having used term in detailed description or claim to a certain extent
" comprising ", " having ", " having ", " having " or its variant, so that such term is intended to be similar to the side of term "comprising"
Formula is inclusiveness.
Claims (21)
1. a kind of method for being sorted out to the article being placed in object, comprising:
Receive the 3-D image of the article;
The 3-D image of the article is divided into one group of subregion;
For the first subregion in this group of subregion:
Computer tomography (CT) value based on each voxel being arranged in first subregion will be arranged in described
The voxel binning in one subregion is to first group of histogram in;And
When the number of the voxel in first subregion and in the range of known CT value of the CT value in first kind article
When more than specified threshold, the article is classified as potential first kind article.
2. according to the method described in claim 1, wherein, the first kind article corresponds to potential explosive, and institute
State the range of known CT value of the range of known CT value corresponding to explosive.
3. according to the method described in claim 1, wherein, CT value corresponds to density value.
4. according to the method described in claim 1, wherein, CT value corresponds to z virtual value.
5. according to the method described in claim 1, including:
Receive the 3-D image for being placed with the object of the article;And
The 3-D image for dividing the object, by the rest part of the 3-D image of the article and the object
3-D image separate.
6. according to the method described in claim 5, including:
The feature vector of the article is determined using the 3-D image of the article;And
Region defined below: the region includes the 3-D image of the article and the 3-D image with the article
One group of voxel that is adjacent and not representing the article.
7. according to the method described in claim 6, wherein, the division includes the 3-D image that will include the article
The region division is one group of subregion.
8. according to the method described in claim 6, wherein, the 3-D image for receiving the article includes reception and institute
State this group of adjacent voxel of the 3-D image of article.
9. according to the method described in claim 1, including:
For the second subregion in this group of subregion:
Based on the CT value for each voxel being arranged in second subregion, will be arranged in second subregion described in
Voxel binning is to first group of histogram in;And
When the body in second subregion and in the range of the known CT value of the CT value in the first kind article
When the number of element is more than the specified threshold, the article is classified as the potential first kind article.
10. according to the method described in claim 9, wherein, regardless of in first subregion and CT value is described first
Whether the number of the voxel in the range of the known CT value of types of articles is more than the specified threshold, as long as described
In second subregion and the number of the voxel of the CT value within the scope of the known CT value of the first kind article is super
The specified threshold is crossed, the article is just classified as the potential first kind article.
11. according to the method described in claim 9, wherein, first subregion and second subregion overlap.
12. according to the method described in claim 9, wherein, first subregion and second subregion are mutual exclusions.
13. according to the method described in claim 9, wherein, first subregion has first thickness, and second son
Region has the second thickness different from the first thickness.
14. according to the method described in claim 1, including:
The feature vector of the article is determined using the 3-D image of the article, wherein each of this group of subregion
Subregion corresponds to the piece along first eigenvector of the article.
15. according to the method described in claim 1, including:
It is calculated based on the CT value for each voxel being arranged in first subregion for first subregion
Histogram.
16. a kind of calculating equipment, comprising:
Processing unit;With
Memory, the memory include processor-executable instruction, and the processor-executable instruction is single by the processing
Member performs the following operation when executing:
The three-dimensional image segmentation of object is indicated at one group of article;
Calculate the article characteristics of this group of article expression;
Based on the article characteristics that the potential hiding article in this group of article expression for potential hiding article indicates, from this group of article
The potential hiding article is identified in expression;
Limit the region in the 3-D image of the object comprising the potential hiding article;
It is one group of subregion by the region division;And
For the first subregion in this group of subregion, in first subregion and CT value is in first kind article
When the number of voxel in the range of known CT value is more than specified threshold, the potential hiding article is classified as including potential
First kind article.
17. calculating equipment according to claim 16, wherein the article characteristics instruction is electronic equipment.
18. calculating equipment according to claim 16, wherein
For the second subregion in this group of subregion, in second subregion and CT value is in the first kind object
When the number of voxel in the range of the known CT value of product is more than the specified threshold, the potential hiding article is sorted out
Being includes the potential first kind article.
19. calculating equipment according to claim 16, wherein the potential hiding article is classified as being included in described
The potential first kind article in one subregion rather than in second subregion.
20. calculating equipment according to claim 16, wherein the potential hiding article is classified as being included in described
The first part of the potential first kind article in one subregion and described potential in second subregion
First kind article second part.
21. a kind of non-transitory computer-readable medium, the non-transitory computer-readable medium includes computer executable instructions,
The computer executable instructions are proceeded as follows when being executed by processing unit:
The three-dimensional image segmentation of object is indicated at one group of article;
Calculate the article characteristics of this group of article expression;
Based on the article characteristics that the potential hiding article in this group of article expression for potential hiding article indicates, from this group of article
The potential hiding article is identified in expression;
Limit the region in the 3-D image of the object comprising the potential hiding article;
It is the first subregion, the second subregion and third subregion by the region division;
The CT value of first subregion, second subregion and the voxel in the third subregion is assessed, to determine the
The example of one type object is present in any one in first subregion, second subregion or the third subregion
In sub-regions;And
When determining that the first kind object is present in first subregion, second subregion or the third subregion
Any one of in subregion when, the potential hiding article is classified as including potential first kind article.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/US2016/057643 WO2018075024A1 (en) | 2016-10-19 | 2016-10-19 | Item classification using localized ct value distribution analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109863499A true CN109863499A (en) | 2019-06-07 |
Family
ID=57249872
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201680090235.1A Pending CN109863499A (en) | 2016-10-19 | 2016-10-19 | Sorted out using the article that local computer tomography Distribution value is analyzed |
Country Status (4)
Country | Link |
---|---|
US (1) | US11087468B2 (en) |
EP (1) | EP3529741A1 (en) |
CN (1) | CN109863499A (en) |
WO (1) | WO2018075024A1 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10977755B2 (en) * | 2017-11-21 | 2021-04-13 | International Business Machines Corporation | Cognitive screening for prohibited items across multiple checkpoints by using context aware spatio-temporal analysis |
US11188800B1 (en) * | 2018-04-03 | 2021-11-30 | Robert Edwin Douglas | Method and apparatus for improved analysis of CT scans of bags |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6195444B1 (en) * | 1999-01-12 | 2001-02-27 | Analogic Corporation | Apparatus and method for detecting concealed objects in computed tomography data |
CN102203801A (en) * | 2008-10-30 | 2011-09-28 | 模拟逻辑有限公司 | Detecting concealed threats |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5712926A (en) * | 1994-10-20 | 1998-01-27 | Eberhard; Jeffrey Wayne | X-ray computed tomography (CT) system for detecting thin objects |
JP3826478B2 (en) | 1997-03-13 | 2006-09-27 | ソニー株式会社 | Headphone device |
US7539337B2 (en) | 2005-07-18 | 2009-05-26 | Analogic Corporation | Method of and system for splitting compound objects in multi-energy computed tomography images |
CA2640884C (en) * | 2006-07-20 | 2010-02-23 | Optosecurity Inc. | Methods and systems for use in security screening, with parallel processing capability |
US8090169B2 (en) * | 2007-12-31 | 2012-01-03 | Morpho Detection, Inc. | System and method for detecting items of interest through mass estimation |
US8260020B2 (en) * | 2007-12-31 | 2012-09-04 | Morpho Detection, Inc. | Image based computed tomography number and volume corrections for thin objects in computed tomography systems |
EP2788956B1 (en) | 2011-11-22 | 2020-09-23 | Smiths Heimann GmbH | Laptop detection |
US10572963B1 (en) * | 2017-07-14 | 2020-02-25 | Synapse Technology Corporation | Detection of items |
-
2016
- 2016-10-19 EP EP16791726.9A patent/EP3529741A1/en active Pending
- 2016-10-19 US US16/342,474 patent/US11087468B2/en active Active
- 2016-10-19 WO PCT/US2016/057643 patent/WO2018075024A1/en unknown
- 2016-10-19 CN CN201680090235.1A patent/CN109863499A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6195444B1 (en) * | 1999-01-12 | 2001-02-27 | Analogic Corporation | Apparatus and method for detecting concealed objects in computed tomography data |
CN1333899A (en) * | 1999-01-12 | 2002-01-30 | 模拟技术公司 | Apparatus and method for detecting concealed objects in computed tomography data |
CN102203801A (en) * | 2008-10-30 | 2011-09-28 | 模拟逻辑有限公司 | Detecting concealed threats |
Also Published As
Publication number | Publication date |
---|---|
US11087468B2 (en) | 2021-08-10 |
US20190259160A1 (en) | 2019-08-22 |
EP3529741A1 (en) | 2019-08-28 |
WO2018075024A1 (en) | 2018-04-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11625827B2 (en) | Visual augmentation of regions within images | |
US7295019B2 (en) | Security scanners with capacitance and magnetic sensor arrays | |
EP2443441B1 (en) | Method and apparatus for assessing the threat status of luggage | |
US9390523B2 (en) | Determination of z-effective value for set of voxels using CT density image and sparse multi-energy data | |
JP6968826B2 (en) | Systems and methods for generating projected images | |
US20240003832A1 (en) | Combined image generation of article under examination and image of test item | |
JP2008504591A (en) | Method and apparatus for recognizing objects in an image | |
US20140010437A1 (en) | Compound object separation | |
US11436735B2 (en) | Three-dimensional object image generation | |
Weisman et al. | Comparison of 11 automated PET segmentation methods in lymphoma | |
US20080123895A1 (en) | Method and system for fast volume cropping of three-dimensional image data | |
US9355502B2 (en) | Synthetic image generation by combining image of object under examination with image of target | |
CN109863499A (en) | Sorted out using the article that local computer tomography Distribution value is analyzed | |
KR20240025683A (en) | Security inspection CT target identification method and device | |
CN105372713A (en) | X-ray diffraction imaging system and method of operating the same | |
US9846935B2 (en) | Segmentation of sheet objects from image generated using radiation imaging modality | |
WO2005036207A2 (en) | Security scanners with capacitance and magnetic sensor arrays |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190607 |